apunkt/GPy
1
0
Fork 0
mirror of https://github.com/SheffieldML/GPy.git synced 2026-05-04 01:02:39 +02:00
Code Issues Projects Releases Packages Wiki Activity Actions

prod now seems to work for sparse

Browse source
This commit is contained in:
James Hensman 2014-02-19 22:23:07 +00:00
parent 92d71384b7
commit de51ad638a
6 changed files with 34 additions and 41 deletions
Download patch file Download diff file Expand all files Collapse all files

13
GPy/core/sparse_gp.py
View file

@ -68,22 +68,21 @@ class SparseGP(GP):
self.posterior, self._log_marginal_likelihood, self.grad_dict = self.inference_method.inference(self.kern, self.X, self.X_variance, self.Z, self.likelihood, self.Y)
self._update_gradients_Z(add=False)
def _raw_predict(self, Xnew, X_variance_new=None, which_parts='all', full_cov=False):
def _raw_predict(self, Xnew, X_variance_new=None, full_cov=False):
"""
Make a prediction for the latent function values
"""
if X_variance_new is None:
Kx = self.kern.K(self.Z, Xnew, which_parts=which_parts)
Kx = self.kern.K(self.Z, Xnew)
mu = np.dot(Kx.T, self.posterior.woodbury_vector)
if full_cov:
Kxx = self.kern.K(Xnew, which_parts=which_parts)
var = Kxx - mdot(Kx.T, self.posterior.woodbury_inv, Kx) # NOTE this won't work for plotting
Kxx = self.kern.K(Xnew)
var = Kxx - mdot(Kx.T, self.posterior.woodbury_inv, Kx)
else:
Kxx = self.kern.Kdiag(Xnew, which_parts=which_parts)
Kxx = self.kern.Kdiag(Xnew)
var = Kxx - np.sum(Kx * np.dot(self.posterior.woodbury_inv, Kx), 0)
else:
# assert which_parts=='all', "swithching out parts of variational kernels is not implemented"
Kx = self.kern.psi1(self.Z, Xnew, X_variance_new) # , which_parts=which_parts) TODO: which_parts
Kx = self.kern.psi1(self.Z, Xnew, X_variance_new)
mu = np.dot(Kx, self.Cpsi1V)
if full_cov:
raise NotImplementedError, "TODO"